Prevalence of MAFLD-Related Hepatocellular Carcinoma

Abstract

We read with great interest the article by Vitale et al.,[1] who examined 6882 patients with hepatocellular carcinoma (HCC) in Italy to estimate the prevalence of metabolic-associated fatty liver disease (MAFLD).[1] They found the prevalence of MAFLD to be increasing rapidly among such patients in Italy. They also found that patients with MAFLD-related HCC were at a lower risk of HCC-related death than patients with other HCC subtypes. We applaud the investigators for providing important insights into the prevalence of MAFLD-related HCC. However, we wish to draw attention to some points to place their findings in perspective. Vitale et al. reported that they diagnosed patients with MAFLD if they were overweight or obese, defined as a body mass index > 25 kg/m2, or had type 2 diabetes mellitus, or showed evidence of metabolic disorders.[1] This is not entirely consistent with international consensus guidelines, according to which MAFLD should be diagnosed based on the presence of fatty liver as well as one of the following: overweight/obesity, type 2 diabetes mellitus, or lean/normal weight with evidence of metabolic disorders.[2] In addition, Vitale et al. apparently did not assess waist circumference, insulin resistance, or levels of C-reactive protein, which should also be taken into account when diagnosing MAFLD. Even if the prevalence reported by Vitale et al. can be taken to reflect bona fide MAFLD, we are concerned about whether it can be extrapolated to the larger population of HCC patients in Italy or elsewhere. Among the 6882 patients whom Vitale et al. enrolled consecutively from January 2002 to December 2019, 17.1% had single-etiology MAFLD and 51.2% had mixed-etiology MAFLD, while 31.6% did not have MAFLD. These proportions differed significantly from the corresponding proportions of 10.9%, 30.1%, and 59.0% from another sample of Italian HCC patients[3] (p<0.001, Fig. 1a) despite the fact that the enrollment window of the second sample lay within that of the sample of Vitale et al., and some patients in both samples came from the same medical center. Vitale et al. combined the prevalences of singleand mixed-etiology MAFLD to give an overall prevalence of MAFLD-related HCC of 68.3%,[1] which is significantly higher than the 12.8%[4] or 19.8%[5] in Mainland China or 38.8% in Taiwan[6] (both p<0.001, Fig. 1a). The latter two prevalences seem more reasonable to us given recent estimates of the global prevalence of MAFLD as 39.2% in the general Yu-Xian Teng,1 Hao-Tian Liu,1 Zhu-Jian Deng,1 Jia-Yong Su,1 Yi-Hua Lu,1 Jian-Hong Zhong1,2